Adaptive Dual-Axis Style-Based Recalibration Network With Class-Wise Statistics Loss for Imbalanced Medical Image Classification
IEEE Transactions on Image Processing · 2025
DOI: 10.1109/tip.2025.3551128 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2025. Verfasst von Xiaoqing Zhang und 8 weiteren Autor:innen. Hauptthemen laut MEDLINE: Neural Networks, Computer, Image Interpretation, Computer-Assisted, Deep Learning, Diagnostic Imaging. Die Arbeit wurde bisher 35-mal zitiert (Crossref) und nennt 76 Literaturangaben. Gefördert von: National Key Research and Development Program of China, Shenzhen Basic Research Fund, Shenzhen Key Laboratory of Intelligent Bioinformatics.
Abstract (Originalsprache)
Salient and small lesions (e.g., microaneurysms on fundus) both play significant roles in real-world disease diagnosis under medical image examinations. Although deep neural networks (DNNs) have achieved promising medical image classification performance, they often have limitations in capturing both salient and small lesion information, restricting performance improvement in imbalanced medical image classification. Recently, with the advent of DNN-based style transfer in medical image generation, the roles of clinical styles have attracted great interest, as they are crucial indicators of lesions. Motivated by this observation, we propose a novel Adaptive Dual-Axis Style-based Recalibration (ADSR) module, leveraging the potential of clinical styles to guide DNNs in effectively learning salient and small lesion information from a dual-axis perspective. ADSR first emphasizes salient lesion information via global style-based adaptation, then captures small lesion information with pixel-wise style-based fusion. We construct an ADSR-Net for imbalanced medical image classification by stacking multiple ADSR modules. Additionally, DNNs typically adopt cross-entropy loss for parameter optimization, which ignores the impacts of class-wise predicted probability distributions. To address this, we introduce a new Class-wise Statistics Loss (CWS) combined with CE to further boost imbalanced medical image classification results. Extensive experiments on five imbalanced medical image datasets demonstrate not only the superiority of ADSR-Net and CWS over state-of-the-art (SOTA) methods but also their improved confidence calibration results. For example, ADSR-Net with the proposed loss significantly outperforms CABNet50 by 21.39% and 27.82% in F1 and B-ACC while reducing 3.31% and 4.57% in ECE and BS on ISIC2018.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Förderung
- National Key Research and Development Program of China
- Shenzhen Basic Research Fund
- Shenzhen Key Laboratory of Intelligent Bioinformatics
- National Natural Science Foundation of China
- Henan Provincial Science and Technology Research
Zitieren
APA
Zhang, X., Xiao, Z., Ma, J., Wu, X., Zhao, J., Zhang, S., Li, R., Pan, Y., & Liu, J. (2025). Adaptive Dual-Axis Style-Based Recalibration Network With Class-Wise Statistics Loss for Imbalanced Medical Image Classification. IEEE Transactions on Image Processing. https://doi.org/10.1109/tip.2025.3551128
Vancouver
Zhang X, Xiao Z, Ma J, Wu X, Zhao J, Zhang S, et al. Adaptive Dual-Axis Style-Based Recalibration Network With Class-Wise Statistics Loss for Imbalanced Medical Image Classification. IEEE Transactions on Image Processing. 2025. doi:10.1109/tip.2025.3551128
BibTeX
@article{Zhang2025,
author = {Zhang, Xiaoqing and Xiao, Zunjie and Ma, Jingzhe and Wu, Xiao and Zhao, Jilu and Zhang, Shuai and Li, Runzhi and Pan, Yi and Liu, Jiang},
title = {Adaptive Dual-Axis Style-Based Recalibration Network With Class-Wise Statistics Loss for Imbalanced Medical Image Classification},
journal = {IEEE Transactions on Image Processing},
year = {2025},
doi = {10.1109/tip.2025.3551128}
}
EUVIMED